This proposed clinical trial has been approved by CTEP and our IRB and we have enrolled now 12 patients. Also, we opened the trial at the Rutgers Cancer Institute of New Jersey. The objectives of the study are as follows: - Primary: To determine if there is an improvement in disease-free survival (DFS) with BCG + PANVAC compared with BCG alone in a phase II study in non-muscle invasive high-grade urothelial carcinoma of the bladder who have failed to respond to intravesical BCG - Secondary: Estimate tumor upgrading and tumor upstaging rates for each study arm. - Secondary: Determine the safety of the combination therapy. - Exploratory/Correlative: Evaluate the immune response generated in each arm. Although the major goal is to improve upon the results of BCG alone clinically, there are several scientific objectives that are secondary endpoints of the trial and are as follows: - PPD results will be used to assess the immunologic response to BCG therapy. This premise is based on a previous report that correlates PPD positivity with improved tumor recurrence-free intervals after BCG treatment (Luftenegger et al, J Urol, Feb 1996, Vol 155, 1483-7). - Urine may be collected at baseline (week 0), BCG instillation #1, BCG instillation #3, week 3, week 8, and at the end of the study. Collection volume will range from 30 to 100 mL with each collection. An attempt will be made to collect urine prior to BCG instillation and 2 hours after BCG drainage, based on previous study demonstrating urinary cytokine changes during these time points (Bisiaux et al, J Urol, Vol 181, April 2009, 1571-80) -In particular to the current study, we would like to study the T-cell infiltration pattern within bladder tumors pre- and post-treatment. Specifically, we will look at: - The expression of CEA and MUC-1 antigens in bladder tumor specimens pre- and post-treatment in both arms - The presence of CD4 and CD8 T cells in bladder tumor specimens pre- and post-treatment in both arms - The presence of regulatory T cells (Tregs) by double staining with FoxP3 and CD4 in bladder tumor specimens pre- and post-treatment in both arms - The presence of myeloid derived suppressor cells (MDSC) in bladder tumor specimens pre- and post-treatment in both arms - At baseline, week 3, week 8, and at the end of the study (between weeks 17-20), blood samples will be obtained via 6 green top tubes. From this blood, PBMCs and sera will be isolated, when appropriate samples are available, for the following immune correlates: Immunologic testing will include: 1. IFN-gamma ELISPOT assays for CD8 T-cell responses specific for CEA, MUC-1, and Brachyury (Limited to patients with HLA-A2 allele) 2. Measure CD4 antigen-specific response to CEA 3. Sera samples will be analyzed for the development of antibodies to CEA, MUC-1 and Brachyury. (Samples from baseline and at end of study) 4. Flow cytometry analysis using 10 markers of all PBMC samples for each of the following cell types: CD4, CD8, Tregs, MDSCs, and NK. 5. Serum for NCA (CEA-CAM) titers will be drawn anytime that a patient exhibits new onset neutropenia (ANC 1000) 6. Immunologic studies will be repeated more frequently if clinically indicated, and any abnormalities potentially related to treatment will be followed until they have resolved, or have been determined to not be treatment-related or until the participant's primary medical care is transferred from the principal investigator. We have now successfully accrued 20+ patients to this trial. We are now gearing up to analyze immune responses generated by BCG alone and BCG+PANVAC. We hope that these responses will demonstrate that the addition of PANVAC is inducing immunologic changes and we hope to correlate these changes with responses in patients.